Users do not recognize the internet as a space from which to obtain information simply but as another world different from real world, and use the internet more and more. In this space, users begin to use other selves representing themselves, namely, avatars. Although IDs that are displayed simply by texts on a chatting window represent users themselves in early internet, realistic avatars such as pictures or three-dimensional (3D) models are used with demanding visual elements with the development of hardware and network infrastructures. A user selects one of several models that are provided in advance by a game developer in real gaming space and uses the selected model as his/her avatar by modifying the selected model in various manners to be suited to his/her favorite.
A system for recommending a make-up may be one of application technologies of using such an avatar.
A conventional system for recommending the make-up paints over an avatar image color corresponding to cosmetics and displays the painted avatar. The conventional make-up recommending system does not consider an interaction between cosmetics and skin at all and thus cannot reproduce color of cosmetics.
In addition, the conventional system for recommending the make-up does not consider changes of color of cosmetics and skin tone depending on lighting even though the color of the cosmetics and the skin tone are seen differently from the original ones depending on the change of the lighting.
Due to the reasons described above, existing simulations do not implement precise make-up simulation but only implement a simple function of applying a single product to all users in the same manner. Moreover, the existing simulations cannot determine user's features so that proper make-up style cannot be recommended but that advertising of their products is only promoted.